In characterizing ferns and flowering plants as calcicole or calcifuge, ecologists have relied almost entirely upon the correlation of soil analyses with estimates of plant frequency. In many parts of the British Isles, this is a difficult and arbitrary distinction which breaks down completely in certain circumstances, e.g. the case of plants growing in a soil with a vertical or horizontal leaching gradient, such that the root system is in contact with both acid soil and calcareous soil. It is important, moreover, to take account of the possibility of race differentiation within the species or of variation in shallow soils over calcareous substrata with respect to their capacity to support calcifuge vegetation. Tansley & Adamson (1926), Hope Simpson (1938) and De Silva (1934) have obtained consistent results in estimations of the degree of calcicoly shown by grassland species on the Chalk and Greensand of Southern England. When the ecology of species recognized as calcicoles or calcifuges by Tansley & Adamson, is examined in grasslands of northern and western areas of the British Isles, many discrepancies appear. The most striking of these is the association of 'calcifuges' with calcicolous species on shallow limestone soils (Webb 1935; Balme 1953). This paper describes some preliminary studies of associations of this kind observed on soils over the carboniferous limestone in Derbyshire. The following working definitions will be used: 1. A calcareous soil is one in which the pH of a paste prepared from the top 3 cm of the profile exceeds 6-5. 2. A calcifuge is a species which, in the majority of its field situations, grows on soils in which the pH of a paste prepared from the top 3 cm of the profile does not exceed 6 5. These definitions assume that for most British soils the pH will provide an adequate index of the extent to which the cation exchange complex is saturated with calcium, pH estimations being more rapid, convenient and reliable for this purpose than analyses for exchangeable Ca++ or for 'free' calcium carbonate. Fig. 1 shows that shallow highly humified soils over limestone may have detectable free carbonate even in horizons that are extremely acid. The pH determinations were carried out on pastes obtained by adding sufficient water to thoroughly mixed samples of the top 3 cm of the soil profile to bring them to field capacity. The depth of 3 cm was chosen somewhat arbitrarily as the lower limit of a zone in which superficial leaching will be manifest and of which the base status will be likely to have the greatest influence on the successful establishment of species entering either by seed or vegetatively. Recent experimental evidence (Rorison 1960) suggests that certain calcicole seedlings show markedly impaired growth before they penetrate to this depth on acid soils. A pH of 6 5 seems to mark a fairly consistent upper limit